Bind session regression operations to the native comparison result

Use the fully qualified KeyLoad.Comparisons.OperationResult in the session
fixture's existing Task return type and remove the ineffective outer alias.
Preserve the native interface, operation flow and all result assertions.

Linux Benchmarks run 38983951798 reported one remaining CS0738 and zero
warnings after the retired time-series fixture was removed. This source
checkpoint addresses that reported name-binding failure; successful rebuilding
and benchmark qualification remain open. No local benchmark verification ran.

Author: KSemenenko

tests/KeyLoad.ComparisonTests/Features/BenchmarkComparisons/UnitContracts/Session/Fixtures/SessionRegressionTarget.cs

@@ -1,7 +1,6 @@
 using System.Collections.Concurrent;
 using System.Runtime.CompilerServices;
 using KeyLoad.Comparisons;
-using OperationResult = KeyLoad.Comparisons.OperationResult;
 
 namespace KeyLoad.UnitTests.Features.BenchmarkComparisons;
 
@@ -81,7 +80,7 @@ private sealed class Session(SessionRegressionTarget target) : IDocumentComparis
     {
         public Guid ClientIdentity { get; } = Guid.NewGuid();
         public bool SingleOperationTiming => true;
-        public async Task<OperationResult> ExecuteAsync(Scenario scenario, BenchmarkDocument document, CancellationToken cancellationToken)
+        public async Task<global::KeyLoad.Comparisons.OperationResult> ExecuteAsync(Scenario scenario, BenchmarkDocument document, CancellationToken cancellationToken)
         {
             if (!target.SynchronousOperations)
             {